Breast cancer is a leading cause of cancer-related deaths in women of all ages. Despite advances on diagnosis and therapy, the rate of mortality remains high. While it is a heterogeneous disease associated with diverse genetic changes, the aggressiveness of breast cancers strongly correlates with the level of differentiation. For example, the high-grade tumors contain poorly differentiated cells and exhibit ES (embryonic stem) like gene expression signature. These tumors relapse with high frequency, they are metastatic and they respond poorly to the available therapies. The goal of this proposal is to investigate the mechanism that leads to high-grade progression. We will test the hypotheses that the transcriptional regulator FOXM1 plays a major causal role in the development of high-grade, aggressive breast cancer, and that inhibition of FOXM1 will impede high-grade progression by inducing tumor differentiation. The hypotheses are based on the following observations, First, FOXM1 is over-expressed in HER2- amplified, ER-, and in the triple negative (ER-, PR- and HER2-) basal-like types of breast cancers. Moreover, there is a strong correlation between poorly differentiated breast cancers and over-expression of FOXM1. In normal mammary gland, over-expression of FOXM1 increases the population of luminal progenitors, while deletion of FoxM1 depletes the progenitor/stem pool and increases the differentiated luminal cells. FOXM1 increases the progenitor pool by inhibiting expression of GATA-3, a master activator of luminal differentiation. Interestingly, FOXM1 inhibits expression of GATA-3 by inducing hypermethylation of the CpG islands in its promoter through recruitment of DNMT3B. Surprisingly, the inhibition of GATA-3 by FOXM1 involves the RB tumor suppressor protein. These observations on luminal differentiation suggest the possibility that FOXM1 plays a direct role in the high-grade progression. Our mechanistic model is that FOXM1 drives accumulation of the poorly differentiated breast cancer cells by repressing GATA-3 and other luminal factors. We will use both cell culture and mouse models to investigate these new functions of FOXM1. Also, we plan to investigate the roles of RB because many drugs, directly or indirectly, cause activation of RB, which, in case of the FOXM1 overexpressing tumors, would support accumulation of the poorly differentiated cells through inhibition of GATA-3. Further, we will investigate the hypothesis that targeting FOXM1 will induce differentiation in high- grade tumors. The aims are: 1. Investigate the roles of FOXM1 overexpression in the accumulation of poorly differentiated breast cancer cells. 2. Investigate how phosphorylation impacts FOXM1's ability to regulate mammary/tumor differentiation. 3. Investigate the role of the FOXM1/RB interaction in the accumulation of poorly differentiated breast cancer cells. Overall, the studies will have significant impact towards an understanding of the molecular mechanism underlying high-grade progression of breast cancers. The notion that over-expressed FOXM1 recruits RB to induce accumulation of poorly differentiated tumor cells is expected to have significant impact in the clinic because many anti-cancer drugs generate active RB, which in FOXM1 over-expressing tumors would generate cells that cause relapse of the disease in an aggressive form.